That was DOM Level 1 (1999). Even level 2 (2000) has this as read-write:
http://www.w3.org/TR/DOM-Level-2-HTML/html.html#ID-94282980
Also keep in mind that with relatively few exceptions, W3C simply
trailed and struggled to capture status quo (or some compromise
representation thereof) back then.
/mz
MZ> http://www.w3.org/TR/DOM-Level-2-HTML/html.html#ID-94282980
Ah, now that makes sense. So my theory goes right down the drain =X
MZ> Also keep in mind that with relatively few exceptions, W3C simply
MZ> trailed and struggled to capture status quo (or some compromise
MZ> representation thereof) back then.
Thanks for your insight!
otherwise, is okay for Game Shows and Reality TV but not for any kind
of security practice. It's not okay that your doctor only read the
medical textbooks. It's not okay that your legislation-drafter only
read about security. But this won't change. It'll get worse. Know why?
Because the people who write the legislation are already legislating
even more of their ilk get hired. Yay for the status quo!
* "right" in this case refers to a collection of experienced-based
best practices backed by anecdotal evidence and the statistics of
small numbers which still may or may not make sense but worked in that
specific implementation.
> >per default over the internet. In a LAN based scenario, we were able to
> >query the Netbios naming service even with full blocking enabled.
>
> The firewall in Tiger, and presumably Leopard, only affects TCP services by
> default (you can enable UDP filtering in the Advanced settings). So no change
> here from the status quo.
Nope -- the behaviour we observed did not depend on the protocol by any
means. For example we were able to connect to a netcat server listening on
a TCP port despite of "Set access to specific services and programs" and
an empty list of allowed services.
